The Effectiveness of The Educare Program in Improving Pro-Social Attitudes to Prevent Bullying in Inclusive Schools

The fundamental challenge in implementing inclusive schools is related to accepting students with special needs by other students. Lack of acceptance and respect for differences often trigger acts of bullying. There are insufficient studies describing that the perpetrator of bullying in school is a student with special needs. This study aims to measure the effectiveness of the educare program in improving students' pro-social attitudes to prevent bullying in inclusive schools. The Educare Program consisted of four lesson plan themes and became a treatment instrument. Single subject research (SSR) was used for the research method, implementing the baseline-intervention-maintenance (AB+M) phase with multiple baselines across subjects. Three students, including a student with a disability and a low pro-social attitude, were the subjects of this study. The data were collected through observation and analyzed through graphical visual inspection focusing on trend, level, latency, and percentage of all non-overlapping data (PAND). The results of the study presented in polygon graphs and PAND showed that the Educare Program is very effective in increasing students' pro-social attitudes. Almost all scores of pro-social attitudes obtained through trend with PAND reached 96.11%.


Introduction
In early 2014, Plan International and the International Center for Research on Women (ICRW) revealed an appalling fact about child abuse in schools, showing that 84% of children in Indonesia have experienced violence in school.This figure was higher than the trend in Asian countries (70%).The research was conducted in five Asian countries: Vietnam, Cambodia, Nepal, Pakistan, and Indonesia.In Indonesia, the data were collected from three cities, Jakarta, Serang, and Banten.The survey was conducted from October 2013 to March 2014, involving nine thousand students aged 12-17 years, teachers, principals, parents, and NGO representatives (Bhatla et al., 2014).The same study was conducted in Surabaya by Chandra and Mulya (2009) with 765 high school students.The results showed that 48.2% of students indicated they had been the victims of bullying, and 45.1% admitted they had been the bullying perpetrators.The victims reported that the bullying occurred in the classroom when there were no teachers (35.9%) and the lesson was in progress (30.2%).Meanwhile, as reported in an article published in the United Nations Educational, Scientific and Cultural Organization (UNESCO) news, nearly a third of teenagers are victims of bullying (Otchet, 2018;Rosario Quiroz et al., 2021).
Education in Indonesia is regulated by the law of the Republic of Indonesia Number 23/2002.It is stated that every child who lives and grows up is protected from discrimination and social status in various circumstances (Yuniardi, 2011).Particularly at the superstructure level, Law No. 8 of 2016 on Persons with Disabilities further strengthens the Indonesian government's serious attention to the rights and protection of people with disabilities.Meanwhile, Canada's inclusive education policy that caters to all students' needs indicates that privileges are given to particular groups of students.However, there is a specific focus on students with disabilities in 'regular classrooms.'The implementation is as problematic as the term.However, all students' outcomes are considered the most positive (Hardy & Woodcock, 2015).
Based on the research on the attitude of regular students toward students with special needs that provides praxis of inclusion schools, it was found that students with special needs experience isolation and frustration despite the advantage of interacting directly with regular students.The limitations of students with special needs in inclusive classes make them vulnerable to bullying perpetrated by regular students.Moreover, children with special needs do not understand that bullying is not good.Therefore, it is highly likely that the percentage of bullying perpetrated by regular students by students with special needs in inclusive schools will increase (Hasanah et al., 2015).
Of all the following research on bullying in schools (Cook et al., 2020;Damayanto et al., 2020;Hasanah et al., 2015;Nambiar et al., 2020;Rosario Quiroz et al., 2021), there is no research revealing that the bully is a student with special needs.Based on this finding, the researchers conducted an observation in one of the inclusive schools in Malang.The observation was carried out in early 2022 when face-to-face learning methods had been allowed as the restriction during the Covid-19 pandemic was lifted.A different finding was found in the observation.Bullying was perpetrated by regular students to students with special needs and by students with special needs to their counterparts and regular students.
The principal and teachers at the inclusive school stated that efforts to decrease and prevent violence and bullying are carried out by giving warnings, advising students while participating in religious activities, and giving ethics lessons.However, bullying and domination by some students, including students with special needs, still prevail.Thus, a solution must be found.The researchers consider the need to provide specific interventions for bullying students to help schools systematically, scientifically, and sustainably.
Several research results on school-based interventions for bullying prevention always recommend that pro-social attitudes and social skills are taught to students (Cook et al., 2020;Gaffney et al., 2019;Swearer et al., 2014;Ttofi & Farrington, 2011).Pro-social attitudes and social skills include pro-social skills, respect, empathy, conflict resolution, tolerance, responsibility, relationship skills, problem-solving skills, and assertiveness (Albayrak et al., 2016).The online school-based Bullying Prevention Program by Timmons-Mitchell et al. (2016) provides six social skills to create healthy friendships to reduce students' indications of bullying.The research conducted by Ttofi and Farrington (2011), which examined the effectiveness of school-based programs to decrease bullying, showed a decrease in bullying by 20-30% and in victims of around 17-20%.Teaching pro-social attitudes and social skills has been proven effective in preventing and decreasing bullying in schools (Albayrak et al., 2016;Timmons-Mitchell et al., 2016).Several research findings show that improving pro-social attitudes and social skills can reduce bullying in traditional and inclusive schools.However, research on implementing pro-social improvement programs in the context of bullying prevention has not yet been carried out using a small n-experimental approach.
These findings have inspired the researchers to conduct a program for improving pro-social attitudes to reduce bullying behavior involving students with special needs as perpetrators.In addition to explaining bullying and its effects, the intervention design must incorporate the importance of a caring attitude and mutual respect.Therefore, the intervention program is called Educare (educate, care, and respect).The research problem formulated in this experimental study is as follows: Is the Educare Program effective for improving pro-social attitudes to prevent bullying in inclusive schools?

Literature Review
Inclusive education has been on the global education agenda for the past three decades, leading to overhauling its pedagogical principles.This form of education arises due to the many assessments of the education system that many adhere to, making many children drop out of school or receive education unequally (Cobenas, 2020).Inclusive education is contextualized in local systems and represents an ongoing struggle to ensure access to a meaningful and equitable education field (Miller et al., 2022).In the praxis of inclusive education, teachers and peers play an important role in school to ensure interaction and a sense of belonging between them.From a particular perspective, teachers can promote positive attitudes that develop social skills to facilitate student interaction (Besic et al., 2020).
To be inclusive in the true sense (accepting differences, responsibilities, caring, and respecting), schools should have an obligation to sensitize the needs of students with disabilities and establish a focus on structural changes in attitudes in the classroom (Nambiar et al., 2020).If the management of classes that accepts students with special needs is not done optimally, it will cause bullying attitudes and behaviors (Damayanto et al., 2020).Based on the definition put forward by the American Psychological Association, bullying is "persistent threatening and aggressive physical behavior or verbal abuse directed toward other people, especially those who are younger, smaller, weaker, or in some other situation of relative disadvantage."Research on regular student attitudes toward bullying behavior in schools that accept autistic children and schools that do not have autism centers was conducted by Cook et al. (2020).Children in schools with autism centers increased feelings of anger, pity, sadness, and shame in response to bullying.In contrast, children in schools that did not accept autistic students showed a less friendly response to bullying, except in response to a story depicting an autistic child ostracized by classmates.The research conducted by Cook et al. also indicates that regular students interacting with students with special needs influences attitudes and sensitivities.
Inclusive education that provides equal opportunities for every student with diverse conditions, including people with disabilities, has vulnerabilities in interaction problems.Friendships and interactions among learners require social skills to respect and help each other.Such social skills are rooted in pro-social attitudes, namely individual attitudes and actions to help others without directly benefiting the helper (Baron & Branscombe, 2011), contrary to one's selfish interests that can benefit others (Kline et al., 2019).In more detail, Eisenberg and Mussen (2009) stated that there are several aspects of pro-social behavior, including: (a) sharing, a willingness to share feelings with others in joy and sorrow, (b) cooperating, a willingness to cooperate with others to achieve a goal, (c) helping, a willingness to help others who are in trouble, (d) act honestly, i.e., a willingness to do something as it is without cheating and (e) charitable, i.e., a willingness to give one's belongings voluntarily to people who are in need.These pro-social aspects will go hand in hand with social skills such as cooperation, assertion, empathy, and self-control, which are highly related to low bullying behavior.On the other hand, if these skills are low in students, the bullying behavior frequency is higher (Jenkins et al., 2016).
Meanwhile, social cognitive theory explains that the environment is important in developing and forming individual behavior.Bandura's social cognitive theory has been updated and expanded into social learning theory and developed by Miller and Dollard as cited in Swearer et al., (2014).Social learning theory says that a person can learn not only through direct instruction but also through observation of the behavior of others and the consequences that come with it.There are several things that a person must do so that the learning process can take place: (a) be in a place at the time of observing the behavior, (b) codify the observed behavior, (c) produce the observed behavior, (d) be motivated to perform the behavior.If the behavior that often appears is negative, the individual in the environment will observe and study the behavior and then imitate it to be internalized.These factors can interact with each other to influence learning; Environmental factors influence behavior, behavior affects the environment, the person (cognitive factor) influences behavior, and so on (reciprocity determinism model).Thus, creating a safe, healthy, and bullying-free school environment would be very beneficial if the observed behavior is positive.This is in line with the results of a study by Volk et al. (2015), which found that bullying behavior has a significant relationship with personality and aggressive behavior.However, bullying is not innate but learned from the surrounding environment.If the environment shows friendly and other positive attitudes, bullying behavior will tend to be absent during adolescence.
The table shows Bandura's stages of observational learning that were used in designing the interventions and adapted to the needs and objectives of this study.Each intervention theme is elaborated in a Lesson Plan (LP) with the following stages: (a) attention, (b) retention, (c) production, and (d) motivation (Bandura, 1986;Swearer et al., 2014).The process of exploring the material that has been previously conveyed by interpreting each process (coding) and turning it (transforming) into new information in the form of images or sentences and rehearsing.
The subject can explain and provide examples of pro-social attitudes consisting of (a) inclusivity, (b) responsibility, (c) care, and (d) respect (previously learned through verbal and visual forms).Examples of pro-social attitudes can be shown through images or sentences.

Production
Imitating or translating symbolic concepts that have been coded into overt behavior.
Practice the learning outcomes and simulation on pro-social attitudes (especially when seeing friends who need help or are bullied) Subjects are able to practice a pro-social attitude (inclusive, responsible, caring and respectful) toward their peers, both normal and with special needs.The practice of prosocial attitudes can be carried out through role-playing guided by the instructor.Motivation Provide feedback and reinforcement for the behavior that has been achieved by students The instructor explains about the positive impacts and advantages of students having pro-social attitudes that can prevent bullying, especially to those who are weaker.
Reinforcement is given to subjects, aiming to increase their motivation so that the knowledge and skills that have been learned in the previous session can be repeated and applied in the subject's daily lives.
Based on several studies related to the background of bullying and several other factors that contribute to bullying behavior in schools, the researchers designed a bullying prevention program called the Educare Program which is expected to be effective in improving pro-social attitudes and preventing bullying behavior.Educare is an acronym for Educate, Care, Respect.The Educare Program comprises three important elements: (a) educate about acceptance and honesty to reduce and prevent bullying in schools, implement the skills taught, create a positive, happy, safe, comfortable, and bullying-free school environment (Educate), (b) instill empathy, mutual respect and the feeling of helping others (Care), and (c) teach mutual respect, understand other people's feelings and show a positive attitude toward differences (Respect).This study aims to examine the effectiveness of the Educare Program in improving prosocial attitudes after a series of interventions for some students in an inclusive school who have low pro-social attitudes.

Research Design
This research is n-small experimental research or single-subject research.The single-subject research experiment focused on the procedures of collecting data from each subject and analyzing them individually (Barger-Anderson et al., 2004;Horner et al., 2005) to know the causal relationship between variables after a given treatment.This study analyzed the effects of treatment on the intervention condition, comparing the condition before and after treatment (Horner et al., 2005;Kratochwill et al., 2013).The applied design was AB+M (Baseline-Intervention plus Maintenance) using multiple baselines across subjects' settings, where each subject had a different duration of baseline session (Barlow & Hersen, 1984;Kratochwill et al., 2013).Considering that the design applied in this study was multiple baselines across subjects, interventions using the Educare program were given to each subject separately.However, each intervention involved three to four students in the class so that the observational learning stage of Bandura could be appropriately applied to support the school's program in preventing bullying.

Research Subjects
Based on the assessment results through observation in the baseline phase, three students with low pro-social attitudes participated as subjects in this single-subject research.All three subjects were studied at the inclusive junior high school of Bhakti Luhur Malang.Two of the subjects were regular students (students without special needs), and the other subject was a student with physical disabilities, missing two legs and using prostheses for his mobility.The profile of the three subjects is as follows:

Research Setting
As previously mentioned, this study implemented multiple baselines across subjects.Therefore, the experiment's setting includes the duration of baseline sessions, intervention sessions, and maintenance sessions for each subject and is shown in Tables 3 and 5.Each subject received a 90-minute intervention every Monday from 11:00 am to 12:30 pm for four weeks.To determine the effects of the intervention, observation was conducted by three observers from Tuesday to Saturday during the research subjects' school breaks and group learning activities.Moreover, three other observers were employed to observe the research subjects in the dormitory where they live during afternoon leisure time, meal time, study time, and evening prayer together.Subject-1 received a seven-week session consisting of the baseline phase (one week), the intervention session (four weeks), and the maintenance session (two weeks).The experiment was conducted from Monday to Saturday.On the other hand, the experiment conducted for subject-2 was the same as that of subject-1, except that the baseline phase was longer (two weeks or twelve days).Meanwhile, the experimental setting for subject-3 in the baseline phase lasted for three weeks.The different duration in the baseline phase shows the setting applied, i.e., multiple baselines across subjects.

Experiment Instrument
The intervention was carried out using the Educare program as the experiment instrument, which was elaborated in the form of a lesson plan consisting of four themes, namely: (a) inclusivity, (b) responsibility, (c) care, and (d) respect.The four themes were designed and developed based on the results of studies on pro-social aspects presented by Eisenberg and Mussen (2009) and social skills published by Jenkins et al. (2016).The lesson plan structure was developed based on Bandura's stages of observational learning (Bandura, 1986;Swearer et al., 2014).This intervention instrument was assessed by six experts in educational psychology, psychologists, and an orthopedagogist and was declared valid with a coefficient of 0.82 based on the Aiken's V test (Aiken, 1985).Experimental instruments in the form of lesson plans for the first theme can be seen in the Appendix.

Data Collection and Analysis
The six observers, three at school and three in the dormitory participating in the research, contributed significantly to the data collection.To carry out the observation, the observers were trained to use the observation scale developed by the researchers based on the pro-social scale of Eisenberg and Mussen (2009) and the social skills of Jenkins et al. (2016).Table 6 shows the indicators of pro-social attitudes that became the assessment focus of this study.The assessment of the subjects' pro-social attitudes was scored using a scale of 1 to 5 from the lowest to the highest level of pro-social attitudes (never, rarely, sometimes, often, always).
The data were presented in a graphical visual inspection.The graph was a conversion of research data in a polygon diagram.This graphical visual analysis was made for each lesson plan (within-conditions and between-conditions), taking the following aspects into account: trend (the tendency in the direction of the data trail), latency (the number of sessions needed to achieve behavior change), and level change (the change in behavior score level), and Percentage of All Non-overlapping Data (PAND) (Scruggs & Mastropieri, 2001;Subasno et al., 2021Subasno et al., , 2022)).PAND is the entire data points in the intervention condition that do not overlap with all data points in the baseline condition, divided by the number of data points in the intervention condition multiplied by 100%.If the score is >90%, then the intervention is considered very effective, 90%≥70% effective, 70%≥50% doubtful, and <50% is not effective (Olive & Franco, 2008).

Subject-1
Subject-1 is the youngest (almost 13 years old) compared to the other two subjects when this study was conducted.Subject-1 went through a baseline phase for one week or six sessions, the shortest duration.It was due to physical abuse committed that was often harmful.Therefore, the immediate intervention was necessary for Subject-1.The results of the research conducted through observation of Subject-1 are presented in Table 7, while the data conversion is in the form of a polygon graph, can be seen in The baseline session for subject-1 lasted for one week (six days).During the observation period of pro-social attitudes that include inclusivity and responsibility, care, and respect, each aspect scored below 40 points.Thus, subject-1 did not meet the criteria to obtain intervention immediately using the Educare Program.In the second week, when the Educare Program started, there was a score increase in all aspects of pro-social attitudes.The score continued to increase until the fifth week at the end of the intervention.Although the score increase in each aspect was volatile, all three aspects showed an exponential score, reaching a range of 80-100.The score increases in pro-social attitudes influenced by the Educare Program occurred on the trend element, showing a gradual score increase.Based on the graph, it can be seen that the absence of level change marked an increase in the score.Likewise, the latency element showed that the intervention did not directly have an impact, nor did it appear.Because the intervention was given every Monday and the remaining days of the week were used for observing the effects of the intervention, there was a score increase on the first day after the intervention.However, on the following days, the scores fluctuated and tended to decrease.Furthermore, compared to the scores recorded in the last intervention, the scores decreased more in maintenance conditions.The decrease in score was not lower than the score in the baseline condition.The three aspects of pro-social attitudes managed to score between 70 to 90.Mathematically, it can be concluded that the range of increase from baseline to maintenance condition is from 20 -35 to 70 -90 (50 -55) points.
The percentage of all non-overlapping data (data points in the intervention condition compared to data points in the baseline condition) is presented as follows: Based on PAND calculation, it can be seen that the overlapping data only occurs in one data point, i.e., in the respect aspect.Compared to the movement of points from session to session in the intervention phase, almost all data points increase.Therefore, the Educare Program is categorized as very effective for subject-1 based on PAND.In addition, it is confirmed by a gradual increase of some data points, especially the aspect of care that reaches the perfect score.

Subject-2
Subject-2 was 15 years old when the study was conducted.He was in a baseline phase for two weeks or 12 sessions.Regarding his action, subject-2 occasionally showed bullying behavior, mostly because he was influenced by subject-1.His anti-social behavior is mostly verbal, such as making fun of and demeaning his friends who are physically or academically weaker.His behavior has made many friends feel inferior and afraid of him.Experimental data for subject-2 are presented in Table 9, and data conversion in the form of a polygon graph can be seen in Figure 2.  Subject-2 received a baseline session for 12 sessions (two weeks).The scores of pro-social attitudes successfully recorded through observation were in the range of 20-45.The range of scores is considered low.Therefore, intervention is needed to improve the pro-social attitudes score for subject-2.In the third week, when the Educare Program started, all three aspects of pro-social attitudes immediately experienced an increase in scores.The scores continued to increase until the sixth week at the end of the intervention.Although the increase in score for each aspect fluctuated as in subject-1, all three aspects showed an exponential score, reaching a range of 75-100.The score increases in pro-social attitudes influenced by the Educare Program occurred on the trend element, showing a gradual score increase.Based on the graph, it can be seen that the absence of a latency element indicated that the intervention did not have a direct effect.There was a noticeable level change or increase in score after being given the fourth intervention.The increase in score was shown by the care aspect, with an increase of 40 points, and the respect aspect, which increased to 35.Because the intervention was given every Monday and the remaining days of the week were used for observing the effects of the intervention, there was a score increase on the first day after the intervention.However, on the following days, the scores fluctuated and tended to decrease.Additionally, in the maintenance condition, the scores experienced a decrease compared to the scores recorded in the last intervention.The decline in score was quite noticeable in the care and the inclusivity-responsibility aspect, which reached the score of 60 and 65, respectively.However, the decreasing scores did not reach the same level as those in the baseline condition.The three aspects of pro-social attitudes managed to score between 60 to 90.Mathematically, it can be concluded that the range of increase from baseline to maintenance condition is from 20 -45 to 60 -90 (40 -45) points.Furthermore, the percentage of all non-overlapping data (data points in the intervention conditions compared to data points in the baseline conditions) is presented in Table 10 below.It is almost similar to what subject-1 achieved.Based on PAND calculation, it is clear that the overlapping data only occurs in one aspect, i.e., in the inclusivity and responsibility.From the scores gained in each session of the intervention phase, it can be seen that there is an increasing trend.Moreover, only one aspect of care can gain the perfect score.Nevertheless, the score increases in pro-social attitudes from 45 at the beginning of the intervention to 90 at the end, showing that the Educare Program is categorized as very effective for subject-2 based on PAND.It is also confirmed by a gradual increase of some data points, especially in care that reaches the perfect one.

Subject-3
Subject-3 is the oldest student in the school.He was 16 years old when this study was conducted.He was given three weeks or 18 sessions of the baseline setting.The decision was made to ensure that his attitudes were real acts of antisocial behavior, not just impressions.Subject-3 is the only research subject with special needs (physical disability / missing both legs).Although he has used prostheses, his mobility is still limited.However, unlike other students with disabilities, subject-3 has high self-confidence and tends to be arrogant.He also has a loud voice that often intimidates his friends.To students with weaker academic performance than him, he often pushes their heads by jokingly saying "stupid".Experimental data collected from subject-3 is presented in Table 11, which is then converted into a graph, as shown in Figure 3 below.The baseline session for subject-3 lasted for three weeks.The score range of pro-social attitudes successfully recorded through observation in baseline conditions was 20-55.The highest score (55) was shown in the aspect of respect.Subject-3 is a student with a physical disability, but his pro-social score is still relatively low.Therefore, the intervention was needed to improve his pro-social attitudes.In the fourth week, when the Educare Program started, the aspects of inclusivity-responsibility and care showed a fluctuating increase in score with an upward trend.However, the aspect of respect with the highest score in the baseline condition decreased.Therefore, Educare Programs have an element of latency in improving pro-social attitudes regarding respect.Overall, the scores continued to increase until the seventh week at the end of the intervention.Despite the fluctuations, all three aspects showed an exponential increase reaching the range of 80-100.Based on the graph, it can be seen that the level of change indicated that the given interventions increased the score of care aspect significantly, reaching an increase of 30 points (from 70 to 100).Because the intervention was given every Monday and the remaining days of the week were used for observing the effects of the intervention, there was a score increase on the first day after the intervention compared to the last day of observation in the previous week.Unfortunately, the increase was almost always followed by a fluctuating decline in score the following days.Furthermore, in the maintenance condition, the score of pro-social attitudes decreased compared to the score recorded in the last intervention.However, the decrease did reach the same level as in the baseline condition.The three aspects of pro-social attitudes managed to score between 60 to 85. Mathematically, it can be concluded that the range of increase from baseline to maintenance condition is from 20 -55 to 60 -90 (40 -35) points.The percentage of all non-overlapping data (data points in the intervention condition compared to data points in the baseline condition) is presented as follows: Regarding data points that do not overlap, subject-3 has the lowest score compared to the other two subjects.It is shown by five data points that are overlapping.Thus, it can be concluded that the Educare Program is effective.However, overall the Educare Program has proven to be very effective in improving the pro-social attitudes of subject-3 (students with special needs), with an average effectiveness of 91.67.It can be seen from the graph that showed an upward trend in the intervention phase, the PAND that was concluded to be very effective, and the graph that showed higher scores than scores in the baseline phase.

Table 11. Results of Pro-Social Attitudes Measurement of Subject
From the research results about intervention using the Educare Program for three subjects, it can be seen from the graphs that there was an upward trend, although there were fluctuations during the process.The increase in score from the baseline phase to the maintenance phase is as follows: subject-1 increased by 50-55 points, subject-2 increased by 40-45 points, and subject-3 increased by 35-40 points.The increase in score can be directly equated with the average percentage, indicating that the Educare Program succeeded in increasing the pro-social attitudes of subject-1 by 52.5%, subject-2 by 42.5%, and subject-3 by 37.5%.

Discussion
This research employed a single-subject experimental approach, implementing the Educare Program that increased the score of pro-social attitudes by 37.5% to 52.5%.The percentage is higher than the percentage achieved in research conducted by Ttofi and Farrington (2011) with the same theme.There is a high possibility that the single-subject experimental approach implemented in this research is one of the reasons for the better results.With this approach, every progress of each subject from session to session can be carefully monitored.In addition, the small number of groups (3-4 students) makes it easier for instructors to control and ensure each student's engagement, especially the research subjects.Implementing the Educare Program to improve the pro-social attitudes of the three study subjects cannot be separated from Bandura's social learning.As stated, bullying behavior can occur because the perpetrator learns from what he sees (Swearer et al., 2014).Therefore, as in the observational learning stage, pro-social attitudes can also be built and trained by observation.The Educare Program in this study teaches the themes of inclusivity, responsibility, care, and respect, which are elaborated into an interactive lesson plan, combined with the observational learning stages: attention, retention, production, and motivation (Bandura, 1986;Swearer et al., 2014) to produce good learning.
The results showed a similar pattern of improved pro-social attitudes among the three subjects.Furthermore, what is more encouraging is that the Educare Program can improve the pro-social attitudes and behaviors of students who tend to be unwelcoming, uncaring, and disrespectful to their peers with special needs.This research could be held at Bhakti Luhur Malang inclusive junior high school because the school is committed to implementing inclusive education.This commitment is shown by an open attitude and a willingness to cooperate with the researchers.These are important traits and can strengthen the view of Nambiar et al. (2020) so that the concept of inclusion can have the meaning it should have, where sensitization of the special needs of students with disabilities is carried out.
Subject-1 and subject-2 in the study are regular students without disabilities.They showed a low pro-social attitude and tended to commit bullying, but their pro-social attitudes were successfully fostered through implementing the Educare Program.It seems necessary to find the motives of bullying.Research related to the motives of bullying committed by adolescents states that adolescents want to be valued highly by their peers.Bullies do not have good social skills, so they commit bullying to gain social acceptance from peers and achieve high social status (Berger & Caravita, 2016).
While subject-3 is a student with physical disabilities (missing both legs), he has high and excessive self-confidence, especially when his achievements in certain subjects surpass some of his peers' achievements.With his loud voice, he often intimidates his friends, particularly those with special needs, jokingly saying "stupid" and even pushing his friend's head.It often happens during break time or when the teacher is absent from the class and does not give students structured tasks that can make them busy.In other words, if the inclusive class is not managed optimally, it can lead to bullying (Damayanto et al., 2020).A new finding in this study is that not all students with special needs always become the victims of bullying, as stated by Hasanah et al. (2015).Students with special needs can also become a perpetrator.It can happen because people learn from their neighborhood or school environment, as Field Swearer et al. (2014) stated.
Based on the research results conducted by Cook et al. (2020), it is stated that there is an increase in pity, sadness, and anger when regular students at schools that have autistic centers witness bullying happen to their fellow students (who have poor social interaction skills, including autistic students).In contrast, from this research conducted in an inclusive school accepting more than one type of special needs, the three students as the research subjects were found to have low pro-social attitudes and behaviors.Thus, intervention was required to improve them.
The results of this research, presented in the form of polygon infographics, described a phenomenon not previously anticipated by the researchers.It was found in the assessment that pro-social attitudes and behaviors do not always follow the theme of intervention in the learning (implementation) plan.For example, in the first intervention, the theme taught was inclusivity (acceptance, honesty).After that, observation and assessment were conducted for these aspects and other themes that had not been given.It turned out that the scores for the whole aspect did not differ much and were volatile in a narrow score range.This phenomenon concurs with the results of a study conducted by Volk et al. (2015) which found that bullying behavior has a significant relationship with personality and aggressive behavior.Therefore, when one aspect gets intervention, it will affect other aspects related to pro-social attitudes and behaviors.

Conclusions
The Educare Program (Educate, Care, and Respect) has proven effective in improving pro-social attitudes to prevent bullying in inclusive schools.The effectiveness rate of the Educare Program for subject-1 and subject-2 is 98.33% (very effective), and for subject-3 is 91.67% (very effective).All of the pro-social attitude scores were obtained through the trend mechanism and showed a gradual increase of 52.5% for subject-1, 42.5% for subject-2, and 37.5% for subject-3.
The Educare program has proven to be very effective for the three research subjects because of the appropriate lesson plans (materials, procedures, and methods) and the implementation of a single-subject experimental approach which ensures careful recording of each subject's progress.

Recommendation
Future researchers are recommended to implement the Educare Program using a single-subject research approach and replicate it to more students with low pro-social attitudes, both regular students and students with disabilities.Prosocial materials in the Educare Program can be adapted to the local culture and wisdom, while procedures other than observational learning can be implemented.
Educators, especially in inclusive schools, are encouraged to use the pro-social attitude assessment format and scale in this study to conduct assessments as a precautionary measure for bullying in schools.If it is found that a low pro-social attitude has the potential to become an act of bullying, then the Educare Program, which has proven to be very effective, can be implemented.Educators are also encouraged to develop an assessment format conducted through observation to ensure accuracy, particularly for assessing attitudes toward religious and ethical subjects.

Limitation
Research that applies the SSR approach has limitations regarding the number of subjects and cases, so generalizations can only be made for the same type of case.Therefore, situations and cases that are not the same may require adjustments in research settings.
Figure 1.Educare Program Experiments of Subject-1 Figure 2. Assessment of Pro-Social Attitudes of Subject-2 Figure 3. Assessment of Pro-Social Attitudes of Subject-3

Table 1 .
Stages of Bandura's Observational Learning

Table 2 .
Research Subjects

Table 3 .
Treatment Setting on Subject-1

Table 4 .
Treatment Setting on Subject-2

Table 5 .
Treatment Setting on Subject-3

Table 8 .
Percentage of All Non-Overlapping Data of Subject-1

Table 9 .
Results of Pro-Social Attitudes Measurement of Subject-2

Table 12 .
Percentage of All Non-Overlapping Data of Subject-3